水文资料误差分布特性以及随机模拟

Error Distribution Characteristics of Hydrological Data and Stochastic Simulation

探讨如何模拟水文测验方式方法创新后的水文资料系列,是进行水文分析计算对水文测验方式方法创新的需求分析的第一步。模拟创新系列的前提条件是认为技术创新方案实施后,由于测验方式的改变或测验次数的减少,使水文资料的测验误差有所增加。首先分析了不同类型水文资料的误差分布特性和引入误差的形式,然后选择典型水文站并运用蒙特卡罗随机模拟方法进行验证,研究结果表明创新后的瞬时流量(如洪峰、枯水流量)系列可以通过在原系列中加入服从正态分布的随机误差进行模拟,创新后的时段总量(洪量、年径流)系列可以通过在原系列中加入服从二项分布的相对误差来模拟。

This paper explored how to simulate hydrological data series which were observed after technical innovation in hydrometry.Due to changes in the patterns or reduce number of observation,the error of hydrological data increase is the prerequisite for simulating hydrological data series after innovation.Firstly,this paper analyzed errors distribution characteristics and errors adding form for different types of hydrological data,selected the typical hydrological stations and used Monte Carlo stochastic simulation method to test.The results show that instantaneous flow innovation series after innovation(e.g.peak,low water flow) and total volume series after innovation(e.g.flood volume,annual runoff) can be simulated by adding errors following a normal and binomial distribution into in the original series separately.